Abundance and Run Timing of Adult Pacific Salmon in Big Creek, Becharof National Wildlife Refuge, 2004 Alaska Fisheries Data Series Number 2005–12

U.S. Fish & Wildlife Service

King Salmon Fish and Wildlife Field Office King Salmon, Alaska September 2005 The Alaska Region Fisheries Program of the U.S. Fish and Wildlife Service conducts fisheries monitoring and population assessment studies throughout many areas of Alaska. Dedicated professional staff located in Anchorage, Juneau, Fairbanks, Kenai, and King Salmon Fish and Wildlife Offices and the Anchorage Conservation Genetics Laboratory serve as the core of the Program’s fisheries management study efforts. Administrative and technical support is provided by staff in the Anchorage Regional Office. Our program works closely with the Alaska Department of Fish and Game and other partners to conserve and restore Alaska’s fish populations and aquatic habitats. Additional information about the Fisheries Program and work conducted by our field offices can be obtained at:

http://alaska.fws.gov/fisheries/index.htm

The Alaska Region Fisheries Program reports its study findings through two regional publication series. The Alaska Fisheries Data Series was established to provide timely dissemination of data to local managers and for inclusion in agency databases. The Alaska Fisheries Technical Reports publishes scientific findings from single and multi-year studies that have undergone more extensive peer review and statistical testing. Additionally, some study results are published in a variety of professional fisheries journals.

Disclaimer: The use of trade names of commercial products in this report does not constitute endorsement or recommendation for use by the federal government. Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service Abundance and Run Timing of Adult Pacific Salmon in Big Creek, Becharof National Wildlife Refuge, 2004 Jeffry L. Anderson Abstract The King Salmon Fish and Wildlife Field Office operated a weir on Big Creek, Becharof National Wildlife Refuge, in 2004 to estimate Pacific salmon escapement. Salmon are an important Refuge resource, and accurate escapement estimates are needed to conserve these species. In 2004, 11,906 Chinook Oncorhynchus tshawytscha, 24,957 chum O. keta, 10,451 coho O. kisutch, 3,295 pink O. gorbuscha, and 189 sockeye O. nerka salmon passed the Big Creek weir. Chinook and chum salmon migrated past the weir from late June to early September with peaks during the second week of July for Chinook and mid August for chum salmon. Coho salmon first passed the weir in early August with peaks in late August, early and mid September. Age 1.3 Chinook, 0.3 chum, and 2.1 coho salmon were the dominant age classes. Fewer age 1.1 jack Chinook salmon were observed compared to previous years. The sex ratios of Chinook and chum salmon were approximately 1:1, while the coho salmon run was comprised mostly of males (72%).

Introduction The Alaska National Interest Lands Conservation Act specifically mandates that fish populations and their habitats be conserved in their natural diversity within the Becharof National Wildlife Refuge (Refuge; USFWS 1994). The conservation of Chinook Oncorhynchus tshawytscha, chum O. keta, coho O. kisutch, pink O. gorbuscha, and sockeye O. nerka salmon stocks that are exploited in commercial, sport, and subsistence fisheries requires accurate monitoring of escapement. Big Creek, the largest tributary to the Naknek River, provides important spawning habitat for Chinook, chum, and coho salmon. The majority of adult salmon spawning in Big Creek likely occurs within Refuge boundaries. Therefore, it is necessary to determine current levels of escapement to ensure conservation of these Pacific salmon stocks originating from the Refuge. In addition, information on human use and dependence upon Refuge resources is becoming more important as competition and conflict develops between user groups for the same finite resources (USFWS 1994). Monitoring salmon escapement in Big Creek provides managers with stock status data that can be used to manage these fisheries. Management decisions based on this information will benefit commercial, subsistence, and sport users.

Conflicts between Chinook salmon commercial, subsistence, and sport fishermen in the Naknek River drainage have increased in recent years due to an increase in competition for a limited resource. From 1994 to 1998, the average commercial harvest of Chinook salmon was 4,116 (42%), sport harvest averaged 3,710 (38%), and subsistence harvest averaged 2,009 (20%) (Dunaway and Sonnichsen 2001). From 1990 to 2000, the sport harvest of Chinook salmon in the Naknek River averaged 4,028, with a range from 2,927 in 2000 to 5,552 in 1997 (Walker et al. 2003). In an attempt to balance increasing harvest with escapement goals, the Alaska Department of Fish and Game (ADFG) has implemented seasonal commercial and sport fishing closures, as well as harvest and gear restrictions to protect Chinook salmon in the Naknek River drainage. The biological escapement goal for Chinook salmon in the Naknek drainage is 5,000 spawning adults as indexed by aerial surveys (Dunaway and Sonnichsen 2001). This goal is

Author: Jeffry L. Anderson is a fishery biologist with the U.S. Fish and Wildlife Service. He can be contacted at King Salmon Fish & Wildlife Field Office, P.O. Box 277, King Salmon, AK 99613; or [email protected]. Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service based on index counts, and therefore represents a minimum estimate of total escapement. Chinook salmon returning to spawn in Big Creek represent a sizeable portion of the entire Naknek River escapement. From 1990 to 2002, Chinook salmon escapement in Big Creek has averaged 35% of the total escapement into the Naknek River drainage, ranging from 895 to 2,531 (Sands et al. 2003). Numbers of Chinook salmon counted past the Big Creek weir from 2000 to 2003 ranged from 649 to 10,063, although numbers were not complete counts in some years (Whitton 2003; Anderson et al. 2004).

Coho salmon are also an important species targeted by sport fishermen in the Naknek River drainage, with an average sport harvest of 7,020 during 1990 to 2000 (range 3,253 to 10,276; Walker et al. 2003). Currently, the ADFG has not established a biological escapement goal or management objectives, and no monitoring program exists to assess spawning escapement or overall exploitation of coho salmon in the Naknek River drainage (Dunaway and Sonnichsen 2001). Declines seen in some coho salmon runs in the ADFG Southwest Management Area may be due to excessive harvest, but the lack of escapement data prevents managers from verifying this conclusion (Minard et al. 1998). The lack of escapement data has become a major concern, and without this information it is difficult to determine the health of coho salmon stocks and thus determine appropriate escapement and harvest goals. Coho salmon counts past the Big Creek weir from 2000 to 2003 ranged from 806 to 9,600, although counts were not complete in some years (Whitton 2003; Anderson et al. 2004).

The ADFG does not have biological escapement goals or escapement monitoring for chum or pink salmon in the Naknek River drainage. Harvest in the commercial fishery is monitored, however, and 34,481 chum and 24 pink salmon were harvested in the Naknek River commercial fishery in 2003 (Morstad et al. 2004). The average sport harvest of pink salmon in the Naknek River from 1990 to 2000 was 4,897 (range 1,283 to 23,178), and the average chum salmon sport harvest was 374 (range 179 to 607; Walker et al. 2003). Chum salmon passage at the Big Creek weir ranged from 3,240 to 33,943, and pink salmon passage ranged from 15 to 873 during 2000 - 2003 (Whitton 2003; Anderson et al. 2004).

In 2000, the King Salmon Fish and Wildlife Field Office (KSFO) initiated a multi year study on Big Creek to:

1. Enumerate escapement of Chinook, chum, coho, pink, and sockeye salmon in Big Creek from 15 June to 15 October; 2. Describe the run timing of Chinook, chum, coho, pink, and sockeye salmon through the weir; 3. Estimate the weekly age and sex composition of spawning Chinook, chum, and coho salmon in Big Creek, such that simultaneous 90% confidence intervals have a maximum width of 0.20; 4. Estimate the mean length of Chinook, chum, and coho salmon by sex and age; and 5. Characterize current public use on Big Creek and Becharof National Wildlife Refuge lands by conducting a general survey of boaters passing the weir.

Whitton (2003) reports results of the project from 2000 to 2002; Anderson et al. (2004) reports project results from 2003. This report focuses on work accomplished in 2004.

2 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service Study Area Big Creek originates in the mountains south of Brooks Lake in Katmai National Park and flows northwest about 60 km before joining the Naknek River, 6 km east of King Salmon, Alaska (Figure 1). The drainage is comprised of numerous tributaries, small lakes, and ponds and is almost entirely located within Refuge boundaries. Big Creek is a clear water stream that supports five species of Pacific salmon and spawning populations of rainbow trout O. mykiss, Dolly Varden Salvelinus malma, Arctic grayling Thymallus arcticus, and northern pike Esox lucius. Round whitefish Prosopium cylindraceum, longnose sucker Catostomus catostomus, and lamprey Lampetra spp. are also present in Big Creek. The weir site is 35 km upriver from the confluence of Big Creek and the Naknek River. This section of the stream is characterized by glides and riffles flowing over sand, gravel, and small cobble substrate. Willow Salix spp., birch Betula spp., and grasses dominate the riparian zone. Maximum discharge often occurs during spring breakup, but high discharge also occurs during periods of heavy rainfall between late July and October.

Methods Escapement Monitoring KSFO has operated a weir on Big Creek since 2000. In 2002, the fixed-picket weir used in 2000 and 2001 was replaced with a nontraditional resistance-board weir. The new weir was modeled after designs reported by Tobin (1994), with modifications so that it was constructed and installed in a V-shaped formation that directs upstream migrant fish to a trap box and video passage chute positioned at the apex of the V. Weir panels were constructed from 4.6-m lengths of polyvinyl chloride (PVC) electrical conduit pickets separated by 38-mm lengths of PVC. Aircraft cable (3 mm in diameter) was used to string the pickets and PVC spacers together, and clamps were attached to the ends of the cables to create 1.2-m wide panels. Six stringers were installed on each panel to provide rigidity. Panels were attached to an 8-mm diameter cable secured to the stream bottom with duckbill anchors. Adjustable resistance boards, constructed of plywood (6 mm thick) and waterproof Styrofoam, were attached to the downstream end of panels to provide flotation. Five weir panels were modified to allow boats to pass the weir. The V-shaped resistance-board weir was operated from 23 June to 29 September 2004 when it was submerged by high water. During the period of operation, the weir was inspected, cleaned, and maintained daily to insure integrity.

To facilitate fish passage and reduce the number of fish handled at the weir, a video monitoring system was incorporated into the weir. The video monitoring system consisted of an Applied Microvideo M250 underwater video camera mounted in a sealed aluminum video box filled with filtered water treated with an algaecide. Images were collected through a clear safety-glass window fixed to the front of the video box. The distance between the camera lens and the glass window provided separation between upstream migrant fish and the lens, as needed to obtain full frame images of large Chinook salmon. The sealed video box design was developed to allow underwater video equipment to operate in the fall when frequent rains often create turbid water conditions. In turbid water, image quality is maintained as the majority of the distance between fish and the lens is within the filtered water contained in the video box. The video box was attached to a fish passage chute that blocked external light from affecting the camera, providing a controlled lighting environment. A baffle was installed in the fish passage chute that forced fish to pass within 25 cm of the camera box glass to further improve performance during turbid water conditions. The fish passage chute was lit from above and below by four, 4-ft long, 12-V DC underwater fluorescent light fixtures and from the side by two35-W, 12-V DC underwater

3 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

King Salmon N

N a kne k R 1km iv er

Big Creek

fixed picket weir

floating weir

Figure 1. Big Creek study area showing the fixed picket weir site used during 2000 - 2001, and the floating weir site used since September 2001, Becharof National Wildlife Refuge.

halogen lights mounted near the camera. Lights were positioned to minimize backscatter and disperse lighting to eliminate “hot spots” while still providing enough illumination to identify fish. During normal operation, only the halogen lights were used; all lights were used during turbid water conditions.

The underwater video camera was connected to a Sanyo DSR-3000 digital video recorder using a shielded coaxial cable. The digital video recorder processed the video stream using motion detection hardware and software to eliminate blank footage, and digitally recorded footage containing fish (alarms) for later counting and identification. We tested the motion detection by doing live counts for 60 min and noting whether every fish that passed through the field of view triggered an alarm. We then reviewed the recorded alarm footage and compared counts of each species, number of alarms recorded, and time required to review the footage. After the motion detection was verified to be functional, fish counts were made by reviewing motion detection alarm files.

4 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Vertical sliding doors were installed in both the trap box and fish passage chute. When fish were not being collected for biological sampling, the trap box was closed and fish were passed through the video monitoring chute. This allowed fish to pass freely through the weir unless the video chute was closed to capture fish for biological sampling. The gate to the video chute was normally closed from approximately 22:00 hours until 06:30 hours the next morning to conserve power.

Stream discharge (flow; m3/s) was monitored at a site near the Big Creek weir. The site was selected in an area with uniform water depth and velocity to minimize sampling error. Discharge was measured using the methods of Gordon et al. (1992), and water velocity was measured with a Marsh-McBirney model 201 flow meter. Stage height (ft) was measured twice daily from a staff gauge. A stage-discharge relationship was not established at the Big Creek weir in 2004, as only two discharge measurements were taken.

Water temperature was monitored in 2004 with a Hobo® Temp data logger that recorded temperatures every two hours. The data logger was placed in a secure, well-mixed site near the weir. Water temperature was summarized as daily minimum, mean, and maximum.

Age, Sex, and Length Data We collected Chinook, chum, and coho salmon age, sex, and length (ASL) data using a temporally stratified sampling design (Cochran 1977), with statistical weeks defining strata. Samples were collected uniformly throughout the week (Sunday through Saturday). To avoid potential bias caused by the selection or capture of individual fish, all target species within the trap were included in the sample even if the sample size goal for a species was exceeded. Although weir passage was stratified into statistical weeks a priori, strata for the analysis of Pacific salmon biological data at the Big Creek weir were modified following the field season to represent actual weir passage (Table 1).

Maximum weekly sample size goals for chum and coho salmon were established such that simultaneous 90% interval estimates of age composition for each week have maximum widths of 0.20 (Bromaghin 1993; Table 2). Sample sizes obtained using these methods were increased to account for the expected number of unreadable scales. The derivation of maximum sample size goals was based on a multinomial sampling model (sampling with replacement or small samples relative to a large population). For some salmon species, the sample size goal was expected to be a substantial fraction of the passage in some weeks. Therefore, during weeks of low passage when the maximum sample size goal could not be practically obtained, about 20% of the weekly escapement was sampled. This was sufficient to describe the age composition and reduce the number of fish handled at the weir. For sample size determination, age categories were defined as the total age (fresh water and ocean age combined) for all species.

To gather Chinook salmon length data, we mounted two 12-V DC underwater lasers (635 nm, 10 mW) with line generating lenses below the video camera to measure fish lengths electronically. Lasers were mounted so they projected parallel vertical lines with 100-mm spacing (Figure 2). An onscreen micrometer, (Video Ruler Version 7 DVD; C-Map Systems, Inc. 2004), was then used to measure lengths from captured still images of fish. Prior to using the onscreen micrometer to measure fish lengths, the system was calibrated to correct for barrel distortion near the edges of the camera’s field of view following the procedures of C-Map Systems, Inc. (2004). The onscreen micrometer was calibrated for each captured image by inputting the known 100 mm laser separation distance into the software and then dragging the mouse cursor from one

5 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 1. Strata (time periods) used for analysis of Big Creek biological data, 2004.

Stratum Chinook salmon Chum salmon Coho salmon

1 23 Jun - 3 Jul 23 Jun - 3 Jul 27 Jul - 14 Aug. 2 4 - 10 Jul 4 - 10 Jul 15 - 21 Aug. 3 11 - 17 Jul 11 - 17 Jul 22 - 28 Aug 4 18 - 24 Jul 18 - 24 Jul 29 Aug - 4 Sep 5 25 - 31 Jul. 25 - 31 Jul 5 - 11 Sep 6 1 - 7 Aug 1 - 7 Aug 12 - 18 Sep 7 8 Aug - 20 Sep 8 - 14 Aug 19 - 28 Sep 8 -- 15 - 21 Aug. -- 9 -- 22 - 28 Aug -- 10 -- 29 Aug - 26 Sep --

Table 2. Maximum weekly sample size goals based on the sampling model of Bromaghin (1993) for collecting age, sex, and length data at the Big Creek weir, 2004.

Number of Percent Adjusted Species Age Categories Sample Size Unreadable Sample Size Chum salmon 4 121 15 142 Coho salmon 3 109 15 128

6 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Figure 2. Screen capture of a chum salmon demonstrating use of the Video Ruler 7 DVD onscreen micrometer (C-Map Systems, Inc 2004) used to measure fish lengths electronically in 2004. The vertical red lines are the lasers projected from beneath the camera, and the blue horizontal line is the length measurement.

laser line to the next. Following calibration, the length of the fish was measured by dragging the mouse cursor from the middle of the fish’s eye to the fork of its tail. Only fish that presented a fully extended side-on aspect (Figure 2) were measured. The accuracy of the onscreen micrometer was verified by comparing micrometer-measured fish lengths to those measured in the trap box with calipers. Mid-eye-to-fork length for each fish was measured with calipers and recorded; the same fish was then passed through the video chute and measured with the onscreen micrometer. The relationship between caliper-measured and micrometer-measured fish lengths was determined using simple linear regression. The regression equation was then used to standardize onscreen micrometer lengths to caliper lengths for analysis.

In 2004 we used an age-length key (ALK) to describe the age composition of Chinook salmon at the Big Creek weir. The use of the laser measurement system allowed us to sample lengths on a large number of fish without having to physically trap and handle them. Based on modeling of 2003 data with the equations of Quinn and Deriso (1999), the ALK would produce similar accuracy and precision to the current multinomial sampling design (Table 3). The benefit of using the ALK is that we could produce valid age composition estimates with a much smaller sample size. We chose weekly sample size goals of A = 50 Chinook salmon for age and length

U A l . a

Table 3. Comparison of Chinook salmon weekly age compositions developed using an age-length key (ALK) and the current multinomial S s .

k F

sampling design at the Big Creek weir using 2003 data (Anderson et al. 2004). The ALK was developed from weekly samples of A = 25 a i

s F h i

ˆ s fish for age determination. L = number of fish sampled for length measurement; θ = proportion of fish of age a; CV = coefficient of a

a h n e d r

variation. W e s i

l D

d l a i t f a Age-Length Key (A = 165) e

S S e e r Age 1.1 Age 1.2 Age 1.3 Age 1.4 r v i e i c s

e N

θˆ θˆ θˆ θˆ u Stratum L a SE CV a SE CV a SE CV a SE CV m b 1 97 0.65 0.06 0.10 0.26 0.06 0.23 0.07 0.02 0.33 0.01 0.01 0.53 er 2 0

2 94 0.46 0.06 0.12 0.29 0.05 0.17 0.21 0.04 0.20 0.04 0.02 0.41 0 5 - 3 100 0.33 0.05 0.16 0.30 0.05 0.18 0.29 0.05 0.16 0.08 0.02 0.30 12 4 45 0.07 0.04 0.48 0.15 0.04 0.27 0.50 0.06 0.13 0.26 0.06 0.22 , Sept

5 41 0.12 0.05 0.38 0.16 0.04 0.27 0.41 0.07 0.17 0.29 0.07 0.23 e m be

8 6 44 0.35 0.07 0.21 0.24 0.06 0.25 0.25 0.05 0.21 0.14 0.04 0.29 r 20 0

Multinomial Sampling Design 5 Age 1.1 Age 1.2 Age 1.3 Age 1.4 ˆ ˆ ˆ ˆ Stratum n θ a SE CV θ a SE CV θ a SE CV θ a SE CV 1 122 0.63 0.04 0.06 0.26 0.04 0.15 0.08 0.02 0.25 0.02 0.01 0.50 2 122 0.48 0.04 0.08 0.31 0.04 0.13 0.18 0.03 0.17 0.03 0.02 0.67 3 127 0.25 0.04 0.16 0.27 0.04 0.15 0.31 0.04 0.13 0.17 0.03 0.18 4 75 0.05 0.03 0.60 0.16 0.04 0.25 0.57 0.06 0.11 0.21 0.05 0.24 5 68 0.09 0.03 0.33 0.16 0.04 0.25 0.46 0.06 0.13 0.28 0.05 0.18 6 70 0.39 0.06 0.15 0.20 0.05 0.25 0.33 0.05 0.15 0.09 0.03 0.33

Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

samples collected in the trap box, and L = 200 length samples measured with the laser system such that the coefficient of variation for age 1.4 fish (worst case scenario) would be less than 20%. All age and length samples were used to construct the ALK, and the weekly length samples were used to estimate weekly age composition.

Samples for ASL and ALK data were collected using a dip net to remove fish from the trap box at least once daily or more often as the number of fish moving through the weir increased. Adult salmon were measured to the nearest mm (mid-eye-to-fork length) and the sex of the fish was determined from secondary characteristics. Three scales from each Chinook and coho salmon, and one scale from each chum salmon were removed from the preferred area on the left side of the fish (Jearld 1983), cleaned, and mounted on gummed scale cards. Chum and coho salmon scales were pressed and aged following the field season by KSFO personnel. Standards and guidelines of Mosher (1968) were used in aging scales. Chinook salmon scales were pressed and aged following the field season by ADFG personnel. Salmon ages are reported according to the European method described by Jearld (1983) and Mosher (1968), where the number of winters the fish spent in fresh water and in the ocean is separated by a decimal. Fish with scales that could not be aged, or where sex could not be determined from secondary characteristics, were not included in the ASL analysis. Non-target fishes captured in the trap box were identified to species, enumerated, and released above the weir. Fish were not allowed to hold downstream of the weir. If this occurred, the trap box was closed and the video chute was opened to facilitate upstream passage.

Characteristics of Chinook, chum, and coho salmon passing through the weir were estimated using standard stratified random sampling estimators (Cochran 1977). Within a given stratum m, the proportion of species i passing the weir that are of sex j and age k (pijkm) was estimated as

nijkm pˆ ijkm = , ni++m

where nijkm denotes the number of fish of species i, sex j, and age k sampled during stratum m and a subscript of "+" represents summation over all possible values of the corresponding variable, e.g., ni++m denotes the total number of fish of species i sampled in stratum m. The variance of

pˆ ijkm was estimated as

⎛ n ⎞ pˆ (1− pˆ ) ˆ ˆ ⎜ i++m ⎟ ijkm ijkm v( pijkm ) = ⎜1− ⎟ , ⎝ N i++m ⎠ ni++m −1 where Ni++m denotes the total number of species i fish passing the weir in stratum m. The ˆ estimated number of fish of species i, sex j, age k passing the weir in stratum m ( N ijkm ) was

ˆ N ijkm = N i++m pˆ ijkm , with estimated variance

ˆ 2 vˆ(N ijkm ) = N i++mvˆ( pˆ ijkm ) . Estimates of proportions for the entire period of weir operation were computed as weighted sums of the stratum estimates, i.e.,

9 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

⎛ N ⎞ ˆ ⎜ i++m ⎟ ˆ pijk = ∑⎜ ⎟ pijkm , m ⎝ N i+++ ⎠ and

2 ⎛ N ⎞ ˆ ˆ ⎜ i++m ⎟ ˆ ˆ . v()pijk = ∑⎜ ⎟ v()pijkm m ⎝ N i+++ ⎠

The total number of fish in a species, sex, and age category passing the weir during the entire period of operation was estimated as

ˆ ˆ N ijk = ∑ N ijkm , m with estimated variance

ˆ ˆ ˆ ˆ v(N ijk )= ∑v(N ijkm ). m

If the length of fish of species i, sex j, and age k sampled in stratum m is denoted xijkm, the sample mean length of fish of species i, sex j, and age k within stratum m was calculated as

∑ xijkm xijkm = , nijkm

2 with corresponding sample variance sijkm

⎛ n ⎞ (x − x )2 s 2 = ⎜1− ijkm ⎟ ∑ ijkm ijkm . ijkm ⎜ ˆ ⎟ n −1 ⎝ Nijkm ⎠ ijkm ˆ The mean length of all fish of species i, sex j, and age k ( xijk ) was estimated as a weighted sum of the stratum means, i.e.,

⎛ Nˆ ⎞ xˆ = ⎜ ijkm ⎟x . ijk ∑⎜ ˆ ⎟ ijkm m ⎝ Nijk ⎠ ˆ An approximate estimator of the variance of xijk was obtained using the delta method (Seber 1982),

⎧ 2 ⎫ ⎡ ⎤ 2 ⎪ ⎢ ⎥ ⎛ ⎞ ⎪ ⎪ x Nˆ ⎜ Nˆ ⎟ ⎪ vˆ xˆ = vˆ Nˆ ⎢ ijkm − ijky x ⎥ + ijkm s 2 . ()ijk ∑ ⎨ ()ijkm ⎢ ∑ 2 ijky ⎥ ⎜ ⎟ ijkm ⎬ m Nˆ y Nˆ ⎪ ⎢∑ ijkx ⎛ ˆ ⎞ ⎥ ⎜ ∑ ijkx ⎟ ⎪ x ⎜ N ijkx ⎟ ⎝ x ⎠ ⎪ ⎢ ∑ ⎥ ⎪ ⎩ ⎣ ⎝ x ⎠ ⎦ ⎭

10 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Age characteristics of Chinook salmon were also estimated using an ALK following the methods of Quinn and Deriso (1999). Length (mid-eye to fork of tail) was measured from a simple random sample of size L from the total weir passage, C, as the first stage sample. The sample (L) was classified into 50-mm length intervals with frequencies {Ll}, l = 1,..., J. Scales were collected and ages determined on a sample of size A, with Al taken randomly from Ll using proportional allocation, where the number of fish aged is selected proportional to the length frequencies. Two different methods were used to measure lengths. Most length measurements were taken using the calibrated onscreen micrometer on captured images of Chinook salmon passing through the video chute. Lengths of fish captured in the trap box for the age sub sample were measured with calipers. All length measurements were standardized to caliper length for analysis.

After ages were obtained, age frequencies were classified by length l and age a into an ALK, denoted {Ala}, l = 1,..., J; a = r,..., A where r is the youngest age and A is the oldest age. An

unbiased estimator of the proportion of fish of length l (αˆ l ) was estimated as

L αˆ = l , l L ˆ and an unbiased estimator of the proportion of fish of length l and age a (θ la ) was

ˆ Ala θ la = . Al ˆ An unbiased estimate of the proportion of age a fish (θ a ) was estimated as

ˆ ˆ ˆ θ a = ∑α lθ la , l ˆ with variance ( var(θa ) ) given by

2 ˆ 2 ˆ ˆ ˆ ˆ ˆ ˆ α l θ la (1−θla ) α l (θla −θ a ) var(θ a ) = ∑∑+ . llAl −1 L ˆ The estimate of the total number of age a Chinook salmon in the population (Ca ) was estimated as

ˆ ˆ Ca = Cθ a with estimated variance

ˆ 2 ˆ var(Ca )= C var(θ a ).

Public Use Survey During 2004, boats traveling past the weir were interviewed opportunistically. The following information was obtained for each interview: (1) primary purpose for visit (hunting, fishing, other); (2) secondary purpose (hunting, fishing, other); (3) reason (subsistence, sport, or other);

11 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

(4) residence (city, state, or country of residency); (5) guiding status (guided or unguided); (6) target species and the number kept; (7) group size; and (8) time spent on the Refuge (hours or days). Unless boat passage occurred when the crew was present at the weir, boaters were not likely to be interviewed as the weir design allows boats to motor over the weir without stopping.

Public Outreach A microwave link was installed at the Big Creek weir in 2004 that allowed us to broadcast a live image from the underwater camera back to a receiver site at the U. S. Fish and Wildlife Service Visitor Center in King Salmon. A pan-tilt-zoom camera was also installed on a tripod tower at the weir site that was remotely controlled at the Visitor Center. Visitors could switch between views from the underwater and overhead cameras, and the video signal was displayed on a 32” diagonal flat panel screen mounted on a wall. Video and audio connections were also available, enabling visitors to record video from the system to a personal camcorder. A second microwave receiver was installed at KSFO, but we did not have the ability to switch between camera views; we received the video signal from whichever camera the Visitor Center had selected.

Results Escapement Monitoring We counted 11,906 Chinook, 24,957 chum, 10,451 coho, 3,295 pink, and 189 sockeye salmon past the Big Creek weir from 23 June to 29 September 2004 (Figures 3 and 4; Appendix A). Chinook salmon were observed at the weir from 23 June to 20 September, with a peak of 1,770 fish passing the weir on 9 July. Chum salmon passed the weir from 23 June to 26 September, with a peak of 1,732 on 8 August. Coho salmon were counted at the weir from 27 July to 28 September, with a peak of 2,641 on 20 September. Estimates for coho salmon are incomplete, as the weir was inundated with high water and not fish tight for several days in August and September (Table 4); the weir was not fish tight after 20 September, although we were able to operate the video equipment and count fish until 29 September. Pink salmon migration peaked in early August, and sockeye salmon had no defined peak migration past the Big Creek weir in 2004 (Figure 4). Dolly Varden (N = 1,529), rainbow trout (N = 673), round whitefish (N = 179), Arctic grayling (N = 55), northern pike (N = 14), and lamprey (N = 3) were observed passing through the weir in 2004 (Figures 4 and 5; Appendix B). Most Dolly Varden passed the Big Creek weir in early July 2004 (Figure 4).

Tests of the motion detection were conducted on 28 occasions in 2004. The majority of motion detection failures occurred when small resident fish and juvenile salmon failed to trigger alarm events. The motion detection capabilities failed on several occasions during late July and early August when large groups of chum salmon moved through the camera field of view in a continuous stream. During these times, the crew recorded all fish passage manually and reverted to motion detection alarm recording when the chum salmon passage rate slowed.

We were able to compare live counts to motion detection counts on 18 of the 28 occasions when valid start and stop times were recorded for the live and file review counts. Counts of all species were similar with both methods (Table 5). In general, the crew had a tendency to misidentify Chinook salmon jacks, pink salmon, and sockeye salmon during the live counts. It was also difficult to distinguish between rainbow trout and Dolly Varden during live counts when they moved quickly through the field of view or in large groups. The average fish passage rate during the 18 trials was 169 fish/hr (range 5 to 477), and the average number of motion detection alarms recorded was 49 (range 22 to 96).

12 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

2,800 Chinook salmon ( N = 11,906) 2,100

1,400

700

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

2,800 Chum salmon ( N = 24,957) 2,100

e g a s

s 1,400 Pa r i

We 700

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

2,800 Coho salmon ( N = 10,451) 2,100

1,400

700

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

Date

Figure 3. Daily passage of Chinook, chum, and coho salmon at the Big Creek weir, 2004.

13 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

240 Pink salmon ( N = 3,295) 200

160

120

0 22-Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

240 Sockeye salmon ( N = 189) 200

160

e g a s

s 120 Pa r i 80 We 40

0 22-Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

240 Dolly Varden ( N = 1,529) 200

160

120

0 22-Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

Date

Figure 4. Daily passage of pink salmon, sockeye salmon, and Dolly Varden at the Big Creek weir, 2004.

14 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 4. Summary of Big Creek weir failures, 2004.

Date/Time Failed Date/Time Repaired Hours Not Fish Tight

25 June, 05:00 26 June, 08:00 27 27 August, 06:00 29 August, 15:00 57 3 September, 15:00 7 September, 12:00 93 20 September, 21:00 -- --

15 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

30 Rainbow trout ( N = 673)

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

30 Round whitefish ( N = 179)

g 20 ssa

a 15 P r i e 10 W

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

35 30 Arctic grayling ( N = 55) 25

0 22 -Jun 6-Jul 20-Jul 3-Aug 17-Aug 31-Aug 14-Sep 28-Sep

Date

Figure 5. Daily passage of rainbow trout, round whitefish, and Arctic grayling at the Big Creek weir, 2004.

16 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 5. Comparison of real time versus file review counts by species for n = 18 trials at the Big Creek weir, 2004.

Count Method Species Real Time File Review

Chinook salmon 1,366 1,376 Chum salmon 1,629 1,598 Pink salmon 274 281 Sockeye salmon 17 19 Coho salmon 45 43 Dolly Varden 336 336 Rainbow trout 28 31 Arctic grayling 3 3 Round whitefish 1 1 Other 5 3 Total 3,704 3,691

Discharge measured near the Big Creek weir was 11.4 m3/s (stage height = 1.78 ft) on 4 August and 6.1 m3/s (stage height = 1.39 ft) on 18 August. Peak stage heights in late June, late August, early September, and after 20 September resulted in weir failures in 2004 (Figure 6, Table 4). A stage-discharge relationship was not established at the Big Creek weir in 2004, as only two discharge measurements were taken.

Water temperatures at the Big Creek weir peaked in early July 2004, with a maximum recorded temperature of 21.7°C on 11 July (Figure 7). Temperatures gradually decreased after mid August until the thermograph was removed on 30 September (Figure 7).

Age, Sex, and Length Data Age, sex, and length data were collected from 288 Chinook salmon from 25 June to 2 September, 1,181 chum salmon from 25 June to 18 September, and 256 coho salmon from 2 August to 19 September 2004. Scales could not be collected from two Chinook, 51 chum, and one coho salmon. Scale samples were unreadable or regenerated for 15 Chinook (5%), 86 chum (8%), and 43 coho salmon (17%). Sex could not be determined from external characteristics for 13 Chinook, two chum, and two coho salmon.

Five age classes were identified from Chinook salmon scale samples in 2004. Over all strata, age 1.3 fish comprised the majority of the run (54%, Table 6). Ages 1.2 (29%) and 1.4 (12%) Chinook salmon were also abundant, while only 12 age 1.1 and three age 1.5 fish were sampled at the weir in 2004. Over all strata in 2004, 47% of the Chinook salmon sampled were females (Table 7). Sex composition varied by sample period and ranged from 32% females in strata 1

17 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

5.00

4.50

4.00

3.50

) 3.00 (ft

t h g i 2.50 ff He a

t 2.00 S

1.50

1.00

0.50

0.00 24-Jun 4-Jul 14-Jul 24-Jul 3-Aug 13-Aug 23-Aug 2-Sep 12-Sep 22-Sep 2-Oct

Date

Figure 6. Maximum daily stage height measured at the Big Creek weir, 2004.

18 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

16 ) C ° e ( r

u 12 t a er p m e

T 8

Minimum Mean 4 Maximum

0 30-Jun 10-Jul 20-Jul 30-Jul 9-Aug 19-Aug 29-Aug 8-Sep 18-Sep 28-Sep Date

Figure 7. Daily minimum, mean, and maximum water temperatures near the Big Creek weir, 2004.

19 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 6. Estimated age composition (%) of Chinook salmon by stratum in Big Creek, 2004.

Age 1.1 1.2 1.3 1.4 1.5 Stratum 1 % 4 38 40 15 4 SE (%) 2.6 6.6 6.7 4.9 2.6 n 2 20 21 8 2 Stratum 2 % 4 19 69 9 -- SE (%) 2.6 5.3 6.3 4.0 -- n 2 10 37 5 -- Stratum 3 % -- 55 33 10 2 SE (%) -- 7.7 7.3 4.5 2.3 n -- 23 14 4 1 Stratum 4 % -- 36 39 25 -- SE (%) -- 8.0 8.1 7.2 -- n -- 13 14 9 -- Stratum 5 % -- 13 62 24 -- SE (%) -- 4.9 7.0 6.2 -- n -- 6 28 11 -- Stratum 6 % 6 31 56 6 -- SE (%) 6.2 11.9 12.7 6.2 -- n 1 5 9 1 -- Stratum 7 % 28 36 32 4 -- SE (%) 9.0 9.6 9.4 3.9 -- n 7 9 8 1 -- Total % 4 29 54 12 1 SE (%) 1.4 3.1 3.5 2.2 0.5 n 12 86 131 39 3

20 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 7. Estimated sex composition, sample size, and escapement of Chinook salmon by stratum in Big Creek, 2004.

Sex Stratum n Female (%) Male (%) SE (%) Escapement

1 53 32 68 6.4 1,661 2 53 47 53 6.9 5,360 3 47 62 38 7.1 1,542 4 37 46 54 8.2 1,100 5 50 32 68 6.4 562 6 15 53 47 13.2 975 7 20 55 45 11.3 706 Total 275 47 53 3.7 11,906

and 5 to 62% females in stratum 3. Lengths of Chinook salmon sampled in 2003 ranged from 428 to 935 mm for females, and from 391 to 998 mm for males (Table 8, Figure 8). In general, the more winters a fish spent in the ocean, the larger its size.

We were not able to calibrate the camera and laser system and measure lengths of Chinook salmon electronically until 20 July. Therefore, it was not possible to use an age-length key to generate an unbiased estimate of Chinook salmon age composition for the season as 75% of the run had already passed the weir by that time (Appendix A). However, we were able to verify that the laser measurement system was practical, and an ALK would be suitable to describe Chinook salmon age composition. Once the camera and laser system were calibrated, we measured 78 fish with calipers in the trap box and 64 of the 78 fish with the onscreen micrometer. A simple linear regression model fit the data well (r2 = 0.98, P < 0.00001; Figure 9). From 27 July to 18 August, we captured footage of 685 Chinook salmon passing through the video chute out of a total weir passage of 1,979 (Appendix A). We were able to measure lengths of 598 Chinook salmon with the onscreen micrometer; 87 fish (13%) could not be measured either because the mid-eye to fork length measurement exceeded the camera’s field of view or the fish did not present a side-on aspect to the camera. Most of the Chinook salmon passage we were able to record was from 30 July to 5 August (n = 582). We constructed an ALK using the age data collected from Chinook salmon sampled in the trap box (Al = 271), and length data collected using the onscreen micrometer (standardized to caliper lengths) and measured with calipers in the trap box (Ll = 869; Table 9). Overall estimates of age composition and precision were similar to those obtained with the multinomial sampling design (Table 10). The ALK is biased, however, because the length samples were mainly collected over a 7-d period, and are not representative of the entire run. More large fish were sampled with the lasers during that 7-d period compared to the lengths of fish collected for the age sub-sample throughout the run (Figures 10 and 11), and we did not achieve our goal of proportional allocation for the length sample. The ALK was not used to estimate weekly age composition.

21 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 8. Mean length (mm), SE, range, and sample size by sex and age taken from Chinook salmon at the Big Creek weir, 2004.

Age 1.1 1.2 1.3 1.4 1.5

Female Mean 533 608 780 859 882 SE -- 23.5 43.8 28.4 -- Minimum 428 531 562 726 876 Maximum 584 698 894 935 889 n 2 29 56 26 2

Male Mean -- 618 741 876 -- SE -- 20.9 51.8 42.3 -- Minimum 391 500 551 756 941 Maximum -- 882 960 998 -- n 1 54 75 13 1

Total Mean 426 603 758 863 901 SE 18.4 27.8 48.7 27.4 23.7 Minimum 391 500 551 726 876 Maximum 584 882 960 998 941 n 12 86 131 39 3

22 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

12 Chinook Salmon (n = 288) 10

0 300 400 500 600 700 800 900 1000

12 Chum Salmon (n = 1,178) 10

) 8 % y (

c 6 n ue q

e 4 r F 2

0 300 400 500 600 700 800 900 1000

12 Coho Salmon (n = 256) 10

0 300 400 500 600 700 800 900 1000 Length (mm)

Figure 8. Length-frequency distributions for Chinook, chum, and coho salmon sampled at the Big Creek weir, 2004.

23 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

1000

900 Caliper Length = 12.2755 + 1.0199*Micrometer Length r2 = 0.98, P < 0.00001

800 ) m m

( 700 h t ng e L r 600 pe i l a C 500

400

300 300 400 500 600 700 800 900 1000 Onscreen Micrometer Length (mm)

Figure 9. Relationship between mid-eye to fork lengths measured with calipers and the onscreen micrometer. Solid circles are observed measurements and the solid line is the predicted value.

24 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 9. Age-length key developed from sampling at the Big Creek weir, 2004.

Age

Length (mm) 1.1 1.2 1.3 1.4 1.5 Al Ll

≤ 400 5 0 0 0 0 5 6 401 - 450 6 1 0 0 0 7 17 451 - 500 0 2 0 0 0 2 5 501 - 550 0 10 0 0 0 10 21 551 - 600 1 22 8 0 0 31 49 601 - 650 0 27 11 0 0 38 86 651 - 700 0 17 10 0 0 27 82 701 - 750 0 6 11 1 0 18 66 751 - 800 0 0 29 5 0 34 162 801 - 850 0 0 43 11 0 54 212 851 - 900 0 1 17 10 2 30 130 901 - 950 0 0 1 10 1 12 27 951 - 1000 0 0 1 2 0 3 6 Total 12 86 131 39 3 271 869

25 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 10. Comparison of Chinook salmon age compositions and associated statistics based on multinomial and age-length key (ALK) sampling designs for Chinook salmon at the Big Creek weir, 2004. CV = coefficient of variation.

Age 1.1 1.2 1.3 1.4 1.5

Multinomial Sampling Design % 4 29 54 12 1 SE (%) 1.4 3.1 3.5 2.2 0.5 CV (%) 35.0 10.7 6.5 18.3 50.0

ALK % 3 23 57 16 1 SE (%) 0.6 1.9 2.9 2.4 0.7 CV (%) 23.2 8.4 5.1 14.6 59.3

26 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

100

) Age sub-sample (A = 271) % ( Length sample (L = 598) y 60 nc ue q e r F e v i

t 40 a l u m u C

0 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 Length (mm)

Figure 10. Cumulative length-frequency distributions for Chinook salmon sampled for length (L) with the onscreen micrometer and sub-sampled for age (A) with calipers at the Big Creek weir, 2004.

27 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

10 Length sample (L = 598)

) 0 % 300 400 500 600 700 800 900 1000 y ( nc ue q

e 10 r F Age sub-sample (A = 271) 8

0 300 400 500 600 700 800 900 1000 Length (mm)

Figure 11. Length-frequency distributions for Chinook salmon sampled for length (L) with the onscreen micrometer (top) and sub-sampled for age (A) with calipers (bottom) at the Big Creek weir, 2004.

28 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Four age classes of chum salmon were identified from scale samples in 2004. Age 0.3 fish were the predominant age class sampled in most strata and over all strata (52%; Table 11); age composition varied by sample period in 2004. The sex ratio for chum salmon also varied over sample periods in 2003 and ranged from 30% females in stratum 3 to 69% females in stratum 10 (Table 12). Over all strata in 2004, 52% of the chum salmon sampled at the weir were females. Lengths of chum salmon sampled in 2003 ranged from 456 to 681 mm for females, and from 483 to 768 mm for males (Table 13, Figure 8).

Five age classes of coho salmon were identified from scale samples in 2004. Age 2.1 fish made up the majority of the run in each stratum (except stratum 7) and over all strata (55%, Table 14). One age 1.2, one age 2.2, and five age 3.1 coho salmon were sampled at the Big Creek weir in 2004. The sex ratio for coho salmon also varied over sample periods in 2004, and ranged from 9% females in stratum 5 to 47% females in stratum 2 (Table 15). Over all strata in 2004, only 28% of the coho salmon sampled at the weir were females. Lengths of coho salmon sampled in 2004 ranged from 395 to 633 mm for females, and from 425 to 667 mm for males (Table 16, Figure 8).

Public Use Survey Fifteen groups using the Refuge were surveyed at the Big Creek weir in 2004. Most were hunting and most were from the local area (Table 17). The number of groups interviewed is not representative of total Refuge use during the summer and fall of 2004, as many boats passed the weir without stopping and were not interviewed.

Discussion Numbers of fish of all species observed at the Big Creek weir in 2004 were similar to 2003, and greater than in any previous year from 2000 to 2002 except for chum salmon (Table 18). Although some counts may represent larger escapements of certain species than in previous years (e.g., Chinook, chum, and coho salmon), much of the observed increase in numbers for non-salmon species since 2002 is due to picket spacing and the operation of the video monitoring equipment at the weir. In 2000 and 2001, fish passing through the weir were identified to species and enumerated by an observer looking down into the water column, so water clarity and surface turbulence were factors in accurate observations. In addition, when large numbers of fish were passing through the weir, identification and enumeration focused on Pacific salmon species. Picket spacing in 2000 and 2001 also allowed smaller fish to pass through the weir undetected (Whitton 2003). Picket spacing of the Big Creek weir (38 mm) allows smaller fish (< 300 mm) to pass through without being enumerated, and the size at which fish can no longer pass between the pickets varies by species. In 2003, rainbow trout and Dolly Varden less than 350 mm were not captured in the trap box, although round whitefish, sockeye salmon, and Artic grayling less than this size were captured (Anderson et al. 2004). Picket spacing was adequate to prevent adult Pacific salmon from passing through the weir undetected.

We did not achieve our sample size goal for lengths to estimate the age composition of Chinook salmon using an ALK. Technical difficulties in setting up and calibrating the camera and laser system prevented us from measuring fish lengths electronically until after 20 July. The length samples we did collect after 20 July were not representative of the entire run (Figures 10 and 11), so we were unable to produce weekly age composition estimates. However, the ALK we did produce shows that this method has promise for future years (Table 10). We believe the weekly

29 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 11. Estimated age composition (%) of chum salmon by stratum in Big Creek, 2004.

Age 0.2 0.3 0.4 0.5 Stratum 1 % 7 43 46 4 SE (%) 4.9 9.4 9.5 3.5 n 2 12 13 1 Stratum 2 % 25 55 20 1 SE (%) 3.8 4.4 3.5 0.8 n 30 67 24 1 Stratum 3 % 26 50 24 -- SE (%) 5.1 5.8 4.9 -- n 18 34 16 -- Stratum 4 % 27 58 15 -- SE (%) 3.5 3.8 2.8 -- n 41 88 23 -- Stratum 5 % 23 55 22 -- SE (%) 3.6 4.2 3.5 -- n 30 72 28 -- Stratum 6 % 31 50 18 1 SE (%) 3.9 4.2 3.3 0.7 n 43 69 25 1 Stratum 7 % 39 57 4 -- SE (%) 4.2 4.3 1.7 -- n 52 75 5 -- Stratum 8 % 41 50 8 1 SE (%) 4.4 4.4 2.4 0.8 n 51 61 10 1 Stratum 9 % 49 42 9 -- SE (%) 5.5 5.5 3.2 -- n 37 32 7 -- Stratum 10 % 57 32 11 -- SE (%) 5.1 4.8 3.2 -- n 43 24 8 -- Total % 33 52 14 1 SE (%) 1.7 1.8 1.2 0.3 n 347 534 159 4

30 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 12. Estimated sex composition, sample size, and escapement of chum salmon by stratum in Big Creek, 2004.

Sex Stratum n Female (%) Male (%) SE (%) Escapement

1 34 56 44 8.5 1,221 2 145 43 57 3.9 1,695 3 79 30 70 4.9 668 4 168 39 61 3.6 1,744 5 143 37 63 3.9 1,875 6 149 48 52 4.1 6,539 7 141 60 40 4.1 6,445 8 142 60 40 4.0 3,521 9 95 61 39 4.8 902 10 83 69 31 4.5 347 Total 1,179 52 48 1.7 24,957

31 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 13. Mean length (mm), SE, range, and sample size by sex and age taken from chum salmon at the Big Creek weir, 2004.

Age 0.2 0.3 0.4 0.5

Female Mean 534 560 571 -- SE 10.4 10.8 12.2 -- Minimum 456 472 479 575 Maximum 612 681 634 -- n 186 272 71 1

Male Mean 561 589 607 626 SE 12.5 10.9 15.6 -- Minimum 483 520 510 579 Maximum 630 692 768 646 n 161 261 88 3

Total Mean 547 573 590 615 SE 12.1 12.7 14.2 -- Minimum 456 437 479 575 Maximum 630 692 768 646 n 347 534 159 4

32 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 14. Estimated age composition (%) of coho salmon by stratum in Big Creek, 2004.

Age 1.1 1.2 2.1 2.2 3.1 Stratum 1 % 43 -- 57 -- -- SE (%) 13.5 -- 13.5 -- -- n 6 -- 8 -- -- Stratum 2 % 31 -- 63 6 -- SE (%) 11.8 -- 12.3 6.2 -- n 5 -- 10 1 -- Stratum 3 % 12 -- 76 -- 12 SE (%) 8.0 -- 10.5 -- 8.0 n 2 -- 13 -- 2 Stratum 4 % 28 -- 68 -- 4 SE (%) 6.2 -- 6.4 -- 2.6 n 15 -- 36 -- 2 Stratum 5 % 48 -- 52 -- -- SE (%) 9.3 -- 9.3 -- -- n 14 -- 15 -- -- Stratum 6 % 38 2 58 -- 2 SE (%) 6.9 2.0 7.0 -- 2.0 n 18 1 28 -- 1 Stratum 7 % 57 -- 43 -- -- SE (%) 8.5 -- 8.5 -- -- n 20 -- 15 -- -- Total % 43 < 1 55 < 1 2 SE (%) 4.0 0.2 4.1 0.4 0.8 n 80 1 125 1 5

33 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 15. Estimated sex composition, sample size, and escapement of coho salmon by stratum in Big Creek, 2004.

Sex Stratum n Female (%) Male (%) SE (%) Escapement

1 19 37 63 11.1 487 2 17 47 53 12.3 631 3 21 19 81 8.7 714 4 59 14 86 4.4 2,198 5 32 9 91 5.2 1,325 6 60 22 78 5.2 901 7 46 41 59 7.3 4,195 Total 254 28 72 3.4 10,451

34 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 16. Mean length (mm), SE, range, and sample size by sex and age taken from coho salmon at the Big Creek weir, 2004.

Age 1.1 1.2 2.1 2.2 3.1

Female Mean 544 -- 579 -- -- SE 36.4 -- 22.5 -- -- Minimum 395 -- 443 545 570 Maximum 618 -- 633 -- -- n 23 -- 24 1 1

Male Mean 558 -- 586 -- 625 SE 27.1 -- 22.8 -- 15.3 Minimum 425 552 434 -- 603 Maximum 660 -- 667 -- 646 n 57 1 99 -- 4

Total Mean 553 -- 582 -- 619 SE 29.0 -- 23.3 -- 13.8 Minimum 395 552 409 545 570 Maximum 660 -- 667 -- 646 n 80 1 125 1 5

35 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 17. Summary of public use information collected at the Big Creek weir, 2004.

Category Number of Groups Primary Purpose Hunting 12 Fishing 3

Secondary Purpose Hunting -- Fishing 2 Other --

Hunting Target Species Moose 5 Bear -- Small game 3

Reason for Visit Subsistence 3 Sport 3

Guide Status Guided 1 Not guided 5

Residence King Salmon/Naknek 12 Other 1

36 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 18. Numbers of fish observed at the Big Creek weir, 2000 to 2004. Data for 2000 to 2002 are from Whitton (2003); data for 2003 are from Anderson et al. (2004).

Species 2004 2003 2002 2001 2000

Chinook salmon 11,906 10,063 4,791 649a 1,298 Chum salmon 24,957 33,943 28,812 11,981a 3,241 Coho salmon 10,451a 9,600 806a 4,523 969a Sockeye salmon 189 119 45 38 57 Pink salmon 3,295 873 31 15 80 Dolly Varden 1,529 4,901 347 21b 24 Rainbow trout 673 549 24 11b 2 Arctic grayling 55 83 3 1b 2 Round whitefish 179 210 16 -- 4 Northern pike 14 52 1 -- 2 Longnose sucker ------1

a Incomplete count due to weir failures. b Numbers observed in trap box.

goal of L = 200 length measurements is achievable, as we were able to measure nearly 600 fish over a 7-d period near the end of the run in 2004. We feel confident from this year’s results and modeling of the 2003 data that if enough length samples can be gathered electronically, the ALK will prove to be a viable method to describe the age composition of Chinook salmon. By limiting the required number of fish we need to physically handle to obtain biological data, we can limit potential handling effects on migrating Pacific salmon. One fundamental assumption for this ALK application is that the length at age relationship does not change over the course of the migration, i.e., the length of an age 1.3 Chinook salmon is the same in June as it is in August. We did not collect enough data to test this assumption in 2004.

Chinook salmon age composition in 2004 was similar to that observed in 2000, where age 1.3 fish were the predominant age class (Table 19). The precision (SE) of weekly age and sex estimates for Chinook salmon (Tables 6 and 7) was low compared to previous years due to the smaller weekly sample size goal in 2004 (n = 50) versus goals in previous years (n > 135). The reduced sample size did not allow us to meet the objective for maximum widths of confidence intervals of 0.20 for most ages in most weeks; we did not meet the weekly objective for sex composition precision in any week. However, the age and sex estimates for the entire run were precise. In contrast to 2003, few age 1.1 Chinook salmon (jacks) were observed at the weir, and the sex ratio was nearly 1:1.

37 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 19. Chinook salmon sex and composition (%, standard errors in parentheses) sampled at the Big Creek weir, 2000 to 2004. Data for 2000 to 2002 are from Whitton (2003); data for 2003 are from Anderson et al. (2004).

Age Year Female 1.1 1.2 1.3 1.4 1.5

2000 61 (2.6) 4 (1.1) 34 (2.8) 50 (3.0) 12 (1.8) < 1 2001 50 (3.7) 2 (1.0) 7 (2.3) 33 (4.0) 56 (4.2) < 2 2002 34 (2.5) 18 (2.3) 37 (2.9) 22 (2.4) 22 (2.4) < 2 2003 27 (2.0) 35 (2.0) 25 (2.0) 28 (2.0) 11 (1.0) < 1 2004 47 (3.7) 4 (1.4) 29 (3.1) 54 (3.5) 12 (2.2) < 2

In 2004, the age composition of chum salmon sampled at the Big Creek weir was similar to other even-numbered years in that age 0.3 chum salmon did not dominate the sample (Table 20). A larger proportion of age 0.2 chum salmon was observed than in any other year, and fewer age 0.4 fish were observed than in any other even-numbered year. The proportion of female chum salmon in 2004 was similar to 2003, and higher than most previous years (Table 20).

Age 2.1 coho salmon made up the majority of the run in 2004, but were not as dominant as in previous years; a larger proportion of age 1.1 coho salmon was observed than in any other year (Table 21). Improved freshwater rearing conditions in 2002 may have allowed more age 1 coho salmon to reach smolt size than in previous years. The proportion of female coho salmon sampled was the lowest recorded over the course of the project, although similar to the 2000 return (Table 21).

Peak run timing for Chinook salmon in 2004 (early July, Figure 3) was similar to that in 2003 (Anderson et al. 2004) and 2002, but earlier than the peak in 2000 and 2001 (late July, Whitton 2003). Peak run timing of chum salmon in 2004 (early August, Figure 3) was similar to 2003 (Anderson et al. 2004) and 2000, but later than the peak in 2001 and 2002 (early to mid July, Whitton 2003). Peak run timing for coho salmon in 2004 (mid September, Figure 3) was earlier than the observed peak in 2003 (late September; Anderson et al. 2004) and 2001 (Whitton 2003), the only other years the weir was in operation after early September. However, observed coho salmon migration timing in 2004 is not representative of the entire run due to weir failures (Table 4). The escapement estimate for coho salmon in 2004 is also a minimum estimate, as it is likely that numerous fish passed above the weir without being counted during weir failures in August and September.

38 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Table 20. Chum salmon sex and age composition (%, standard errors in parentheses) sampled at the Big Creek weir, 2000 to 2004. Data for 2000 to 2002 are from Whitton (2003); data from 2003 are from Anderson et al. (2004).

Age Year Female 0.1 0.2 0.3 0.4 0.5

2000 32 (1.7) -- 22 (1.5) 38 (2.0) 38 (1.9) < 2 2001 27 (1.3) -- 2 (0.5) 93 (0.8) 5 (3.1) < 1 2002 43 (1.8) -- 10 (1.0) 28 (2.0) 62 (6.5) < 1 2003 54 (2.0) < 1 3 (0.6) 91 (0.9) 6 (0.6) < 2 2004 52 (1.7) -- 33 (1.7) 52 (1.8) 14 (1.2) < 2

Table 21. Coho salmon sex and age composition (%, standard errors in parentheses) sampled at the Big Creek weir, 2000 to 2004. Data for 2000 to 2002 are from Whitton (2003); data for 2003 are from Anderson et al. (2004).

Agea Year Female 1.1 2.1 3.1

2000 30 (2.6) 9 (1.8) 87 (2.1) 3 (1.1) 2001 50 (1.9) 11 (1.3) 86 (1.4) 2 (0.6) 2002 47 (3.7) 24 (3.6) 72 (3.8) 4 (1.7) 2003 46 (3.0) 16 (2.0) 74 (2.0) 9 (1.0) 2004 28 (3.4) 43 (4.0) 55 (4.1) 2 (0.8)

a Table does not include ages 1.2, 2.2, or 4.1 which were less than 1% of the sample in any given year.

39 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

The use of underwater video monitoring at the Big Creek weir has produced several benefits. Since 2002, recording the images of fish passing through the video chute allowed the crew to pause the images when necessary to accurately identify and enumerate all species passing through the video chute. Reviewing video files captured with the motion detection requires considerably less time than doing live counts, which could result in reduced labor costs to operate the weir. Although the motion detection capabilities failed due to high passage rates of chum salmon on several occasions, counts were not affected as the crew was able manually record fish passage. The video equipment remains functional during high water events when it is not safe or practical to do a live count. For example, the peak coho salmon count in 2004 (2,641 on 20 September; Appendix A) was captured by the video equipment when water levels and velocities prevented the crew from accessing the trap box to do a live count. Although the weir might not be fish tight during these high water events, the data captured by the video equipment would otherwise not be available. The video equipment has also allowed us to obtain more comprehensive counts of resident fish species than in previous years (Table 18). Small rainbow trout, Dolly Varden, and round whitefish that used to swim between the weir pickets or escape from the trap box without being counted are now swimming through the video chute where we can count and identify them. Finally, the use of underwater video technology has removed much of the variability associated with observation conditions. The video chute provides a controlled lighting environment, and the system design allows it to function well in turbid water conditions.

Acknowledgements We thank B. Nielson, B. Blain, M. Rosen, D. Barrett, M. Malick, J. Jensen, S. Shaw, J. Russell, D. Spencer, and C. Dion for their efforts at the Big Creek weir in 2004. We also wish to thank M. Santos for his efforts in designing and constructing the pieces and parts necessary to make the project happen. SeeMore Wildlife Systems provided the technical assistance to set up and operate the microwave link in 2004. Finally, we thank N. Hetrick for his research, dedication, and perseverance in making underwater video monitoring a reality.

40 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

References Anderson, J. L., K. S. Whitton, K. K. Cornum, and T. D. Auth. 2004. Abundance and run timing of adult Pacific salmon in Big Creek, Becharof National Wildlife Refuge, 2003. U. S. Fish and Wildlife Service, King Salmon Fish and Wildlife Field Office, Alaska Fisheries Data Series Report Number 2004-7, King Salmon, Alaska. Bromaghin, J. F. 1993. Sample size determination for interval estimation of multinomial probabilities. The American Statistician 47:203-206. C-Map Systems, Inc. 2004. Video Ruler 7 DVD USB Interactive Manual. Red Lodge, Montana. Cochran, W. G. 1977. Sampling Techniques, third edition. John Wiley & Sons, New York. Dunaway, D. O., and S. Sonnichsen. 2001. Area management report for the recreational fisheries of the southwest Alaska sportfish management area, 1999. Alaska Department of Fish and Game, Fishery Management Report No. 01-6, Anchorage, Alaska. Gordon, N. D., T. A. McMahon, and B. L. Finlayson. 1992. Stream hydrology: an introduction for ecologists. John Wiley & Sons Ltd., Chichester, England. Jearld, A. 1983. Age determination. Pages 301 - 324 in L. A. Nielsen and D. L. Johnson, editors. Fisheries Techniques. American Fisheries Society, Bethesda, MD. Minard, R. E., D. O. Dunaway, and M. J. Jaenicke. 1998. Area management report for the recreational fisheries of the southwest Alaska sportfish management area, 1997. Alaska Department of Fish and Game, Fishery Management Report No. 98-3, Anchorage, Alaska. Morstad, S., and seven other authors. 2004. Annual Management Report, Bristol Bay area 2003. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 2A0-16, Anchorage, Alaska. Mosher, K. H. 1968. Photographic atlas of sockeye salmon scales. Fishery Bulletin 67:243- 280. Quinn, T. J., and R. B. Deriso. 1999. Quantitative fish dynamics. Oxford University Press, New York. Sands, T., S. Morstad, and K. Weiland. 2003. Salmon spawning ground surveys in the Bristol Bay area, Alaska, 2002. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 2A03-07, Anchorage, Alaska. Seber, G. A. F. 1982. The estimation of animal abundance and related parameters, second edition. Macmillan, New York. Tobin, J. H. 1994. Construction and performance of a portable resistance board weir for counting migrating adult salmon in rivers. U. S. Fish and Wildlife Service, Kenai Fishery Resource Office, Alaska Fisheries Technical Report Number 22, Kenai, Alaska. USFWS (U. S. Fish and Wildlife Service). 1994. Alaska Peninsula/Becharof National Wildlife Refuges Fishery Management Plan, King Salmon, Alaska. Walker, R. J., C. Olnes, K. Sundet, A. L. Howe, and A. E. Bingham. 2003. Participation, catch, and harvest in Alaska sport fisheries during 2000. Alaska Department of Fish and Game, Division of Sport Fish, Fishery Data Series No. 03-05, Anchorage, Alaska.

41 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Whitton, K. S. 2003. Abundance and run timing of adult Pacific salmon in Big Creek, Becharof National Wildlife Refuge. U. S. Fish and Wildlife Service, King Salmon Fish and Wildlife Field Office, Alaska Fisheries Technical Report Number 65, King Salmon, Alaska.

42 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Appendix A. Summary of Pacific salmon daily passage at the Big Creek weir, 2004. Chinook salmon less than 508 mm were classified as jacks.

Chinook Chum Coho Pink Sockeye Date Count Jacks Cum Count Cum Count Cum Count Count 22-Jun 0 0 0 0 0 0 0 0 0 23-Jun 4 0 4 1 1 0 0 0 0 24-Jun 5 0 9 6 7 0 0 3 0 25-Jun 6 3 18 48 55 0 0 14 0 26-Jun 0 1 19 15 70 0 0 4 0 27-Jun 77 3 99 171 241 0 0 52 0 28-Jun 56 15 170 164 405 0 0 27 1 29-Jun 620 71 861 213 618 0 0 26 0 30-Jun 27 8 896 29 647 0 0 8 0 1-Jul 607 23 1,526 382 1,029 0 0 55 2 2-Jul 85 4 1,615 91 1,120 0 0 0 0 3-Jul 43 3 1,661 101 1,221 0 0 7 0 4-Jul 265 12 1,938 185 1,406 0 0 17 0 5-Jul 487 16 2,441 339 1,745 0 0 49 0 6-Jul 2 0 2,443 3 1,748 0 0 1 0 7-Jul 402 11 2,856 173 1,921 0 0 56 6 8-Jul 1,577 28 4,461 417 2,338 0 0 129 5 9-Jul 1,728 42 6,231 308 2,646 0 0 167 12 10-Jul 770 20 7,021 270 2,916 0 0 73 4 11-Jul 469 6 7,496 162 3,078 0 0 34 1 12-Jul 902 6 8,404 85 3,163 0 0 42 3 13-Jul 19 2 8,425 32 3,195 0 0 4 7 14-Jul 25 1 8,451 64 3,259 0 0 10 5 15-Jul 70 5 8,526 150 3,409 0 0 0 2 16-Jul 20 0 8,546 113 3,522 0 0 10 1 17-Jul 17 0 8,563 62 3,584 0 0 9 1 18-Jul 135 2 8,700 300 3,884 0 0 71 11 19-Jul 14 1 8,715 335 4,219 0 0 59 6 20-Jul 258 5 8,978 257 4,476 0 0 54 9 21-Jul 394 3 9,375 279 4,755 0 0 102 9 22-Jul 170 2 9,547 227 4,982 0 0 53 2 23-Jul 99 0 9,646 202 5,184 0 0 23 4 24-Jul 17 0 9,663 144 5,328 0 0 9 1 25-Jul 74 1 9,738 305 5,633 0 0 40 1 26-Jul 38 0 9,776 203 5,836 0 0 40 3 27-Jul 23 1 9,800 404 6,240 1 1 131 4 28-Jul 132 4 9,936 210 6,450 0 1 80 5 29-Jul 61 0 9,997 209 6,659 0 1 46 2 30-Jul 114 4 10,115 196 6,855 0 1 68 6 31-Jul 109 1 10,225 348 7,203 0 1 78 4 1-Aug 126 1 10,352 466 7,669 1 2 89 3 2-Aug 249 4 10,605 1,292 8,961 8 10 207 4 3-Aug 262 6 10,873 561 9,522 5 15 94 4 4-Aug 31 2 10,906 640 10,162 17 32 47 1 5-Aug 178 2 11,086 1,491 11,653 40 72 189 1 6-Aug 50 6 11,142 888 12,541 17 89 91 4 7-Aug 52 6 11,200 1,201 13,742 27 116 93 3 8-Aug 125 7 11,332 1,732 15,474 48 164 103 2 9-Aug 52 3 11,387 924 16,398 22 186 62 2 10-Aug 66 7 11,460 1,016 17,414 45 231 82 2 11-Aug 65 2 11,527 1,064 18,478 80 311 66 6 12-Aug 25 2 11,554 855 19,333 88 399 54 5 13-Aug 20 4 11,578 470 19,803 53 452 32 1 14-Aug 9 0 11,587 384 20,187 35 487 19 2 15-Aug 44 2 11,633 890 21,077 124 611 39 6 16-Aug 46 2 11,681 757 21,834 118 729 46 7

43 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Appendix A. continued.

Chinook Chum Coho Pink Sockeye Date Count Jacks Cum Count Cum Count Cum Count Count 17-Aug 27 1 11,709 612 22,446 124 853 45 2 18-Aug 33 13 11,755 470 22,916 106 959 26 2 19-Aug 13 6 11,774 117 23,033 27 986 21 1 20-Aug 12 3 11,789 470 23,503 60 1,046 35 0 21-Aug 2 4 11,795 205 23,708 72 1,118 14 1 22-Aug 2 1 11,798 148 23,856 38 1,156 3 0 23-Aug 11 3 11,812 101 23,957 0 1,156 9 0 24-Aug 2 5 11,819 198 24,155 41 1,197 12 1 25-Aug 12 11 11,842 252 24,407 307 1,504 15 1 26-Aug 4 5 11,851 179 24,586 325 1,829 19 0 27-Aug 0 2 11,853 24 24,610 3 1,832 0 0 28-Aug 0 0 11,853 0 24,610 0 1,832 0 0 29-Aug 5 0 11,858 7 24,617 8 1,840 2 0 30-Aug 7 3 11,868 80 24,697 61 1,901 11 3 31-Aug 2 4 11,874 61 24,758 57 1,958 5 2 1-Sep 7 5 11,886 54 24,812 184 2,142 6 1 2-Sep 6 0 11,892 34 24,846 594 2,736 15 2 3-Sep 4 0 11,896 69 24,915 1,294 4,030 50 1 4-Sep 0 0 11,896 0 24,915 0 4,030 0 0 5-Sep 0 0 11,896 0 24,915 0 4,030 0 0 6-Sep 0 0 11,896 0 24,915 1 4,031 0 0 7-Sep 0 0 11,896 5 24,920 19 4,050 8 0 8-Sep 0 1 11,897 9 24,929 78 4,128 5 0 9-Sep 1 0 11,898 4 24,933 251 4,379 18 0 10-Sep 1 0 11,899 4 24,937 389 4,768 3 0 11-Sep 0 1 11,900 6 24,943 587 5,355 6 0 12-Sep 0 0 11,900 2 24,945 510 5,865 2 0 13-Sep 0 1 11,901 0 24,945 44 5,909 0 0 14-Sep 0 0 11,901 2 24,947 86 5,995 0 1 15-Sep 0 0 11,901 2 24,949 27 6,022 0 0 16-Sep 1 0 11,902 0 24,949 13 6,035 0 0 17-Sep 0 0 11,902 1 24,950 81 6,116 0 0 18-Sep 3 0 11,905 2 24,952 140 6,256 0 0 19-Sep 0 0 11,905 3 24,955 1,459 7,715 0 0 20-Sep 1 0 11,906 0 24,955 2,641 10,356 0 1 21-Sep 0 0 11,906 0 24,955 1 10,357 0 0 22-Sep 0 0 11,906 0 24,955 29 10,386 0 0 23-Sep 0 0 11,906 0 24,955 0 10,386 0 0 24-Sep 0 0 11,906 0 24,955 0 10,386 0 0 25-Sep 0 0 11,906 1 24,956 12 10,398 0 0 26-Sep 0 0 11,906 1 24,957 20 10,418 0 0 27-Sep 0 0 11,906 0 24,957 13 10,431 1 0 28-Sep 0 0 11,906 0 24,957 20 10,451 0 0 29-Sep 0 0 11,906 0 24,957 0 10,451 0 0 Total: 11,477 429 11,906 24,957 10,451 3,295 189

44 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Appendix B. Summary of Dolly Varden (DV), rainbow trout (RT), round whitefish (WF), Arctic grayling (AG), and northern pike (NP) passage at the Big Creek weir, 2004.

Date DV RT WF AG NP 22-Jun 0 5 1 0 0 23-Jun 0 4 7 3 0 24-Jun 2 4 0 1 0 25-Jun 1 1 4 5 0 26-Jun 4 5 2 7 0 27-Jun 11 3 15 0 0 28-Jun 19 7 15 5 2 29-Jun 65 4 5 0 0 30-Jun 32 2 3 0 0 1-Jul 68 28 2 3 0 2-Jul 28 26 3 2 0 3-Jul 17 2 1 1 0 4-Jul 47 3 0 3 0 5-Jul 87 4 2 2 0 6-Jul 21110 7-Jul 232 11 0 2 0 8-Jul 152 5 2 3 0 9-Jul 117 5 7 5 0 10-Jul 89 7 1 0 0 11-Jul 32 2 2 1 0 12-Jul 60 5 1 0 0 13-Jul 19 1 0 0 0 14-Jul 44000 15-Jul 21100 16-Jul 21001 17-Jul 11001 18-Jul 43 1 0 0 0 19-Jul 15 3 0 1 0 20-Jul 38 6 0 0 0 21-Jul 32 6 0 0 0 22-Jul 72001 23-Jul 18 13 2 0 0 24-Jul 21000 25-Jul 26000 26-Jul 11 4 0 0 0 27-Jul 12000 28-Jul 46000 29-Jul 4 14 0 0 0 30-Jul 11 10 0 0 0 31-Jul 99000 1-Aug 117100 2-Aug 197000 3-Aug 512010 4-Aug 116100 5-Aug 813500 6-Aug 411000 7-Aug 523000 8-Aug 912110 9-Aug 56100 10-Aug 826000 11-Aug 313000 12-Aug 312000 13-Aug 214100 14-Aug 27000 15-Aug 19000 16-Aug 322000 17-Aug 516412

45 Alaska Fisheries Data Series Number 2005-12, September 2005 U.S. Fish and Wildlife Service

Appendix B. continued.

Date DV RT WF AG NP 18-Aug 5 12 1 0 1 19-Aug 15 18 1 0 0 20-Aug 3 6100 21-Aug 2 7000 22-Aug 2 4010 23-Aug 3 4000 24-Aug 3 9000 25-Aug 2 6001 26-Aug 1 5100 27-Aug 0 1100 28-Aug 0 1000 29-Aug 0 0000 30-Aug 6 1200 31-Aug 5 6201 1-Sep 2 4100 2-Sep 7 3000 3-Sep 18 11 0 1 0 4-Sep 0 0000 5-Sep 0 0000 6-Sep 0 0000 7-Sep 2 1011 8-Sep 5 8211 9-Sep 3 8000 10-Sep 6 6100 11-Sep 3 6200 12-Sep 1 8200 13-Sep 1 5000 14-Sep 1 5900 15-Sep 1 4400 16-Sep 3 3000 17-Sep 2 3500 18-Sep 5 6400 19-Sep 7 8420 20-Sep 10 30 30 1 1 21-Sep 1 1000 22-Sep 5 6300 23-Sep 0 0000 24-Sep 0 0000 25-Sep 0 8601 26-Sep 2 7000 27-Sep 3 2100 28-Sep 0 7400 29-Sep 0 2100 Total: 1,529 673 179 55 14